In association with remarkable development in the field of computer software, it has become possible to generate on a computer a table document in which a plurality of cells are arrayed in row and column directions and obtain a desired result of spreadsheet computing by giving data such as a computing formula or a numerical value to each cell constituting the table document.
Also in recent years, it has become possible to realize a text-voice synthesizing technology for reading out a text phonetically using a software, and by combining this software for synthesizing voices for a text with the software for spreadsheet computing described above, it is possible to read out a table document on a computer for the purpose of checking or correcting the input data.
To read out the table document, as shown in FIG. 60, a combination of three functions, namely a spreadsheet computing module 90 (spreadsheet computing software), a buffer module 91 (for temporally storing character information), and a text-voice synthesizing module 92 (software for synthesizing voice for a text) are required.
With the spreadsheet computing module 90, which is a first function, by operating the keys a table document is prepared and displayed on the screen. To read a table document, the table document prepared by the spreadsheet computing module 90 is temporally stored as character information for display in the buffer module 91 which is a second function. The table document prepared with the spreadsheet computing module 90 is sent via the buffer module 91 to the text-voice synthesizing module 92. In the text-voice synthesizing module 92, voice synthesis is executed according to the received table document, namely the character information for display. As a result, the table document is read out.
Generally, as only minimum required information is inputted into a cell, additional information such as, for instance, a unit is not appended to the numerical value displayed in each cell, although, on a display screen, it is possible to understand the hidden information such as a unit from a title (item name) of a row or a column.
However, when a table document is phonetically outputted, data in each cell of the table document is sent to the text-voice synthesizing module 92 in a form of character information for display, so that a character string displayed in each cell is read out as it is. For this reason, if a character string is a numerical value, it is difficult to determined whether it is an amount of money, a date, or any other value with a unit, and if checking is executed without understanding the relation between an item and each numerical value, finding out an error becomes more difficult as compared to a case where output on the display screen is checked.
As an analogous technology, the inventions disclosed in Japanese Patent Laid-Open Publication No. HEI 4-20996, Japanese Patent Laid-Open Publication No. HEI 4-24715, and Japanese Patent Laid-Open Publication No. HEI 5-46607 that teaches a better way of understanding the words or numerical values when read out. These publications disclose a technology for phonetically outputting the format information additionally such as an item name or a unit corresponding to a character string such as a word or a numerical value.
In the technology as disclosed in the publications above, when a character string displayed in a cell is read out, an item name or a unit corresponding to the character string is additionally read out, it becomes easier for listeners to understand a item name or a unit (such as a measure) to which a character string belongs, however, as far as an item name is concerned, it becomes odd information when only collation by reading out is required, and this odd information gives the listener an unnecessary load while phonetically understanding the information, and when the character string is a numerical value, the arrangement of figures in the character string or the number of figures can be understood, but if a position, where a unit is inserted into a character is not correct, the numerical value becomes information having a completely different meaning, which makes it difficult to correctly collate a table document by reading out.